I am an Electrical and Computer Engineer based in Thessaloniki, Greece. In 2025 I received my Master of Engineering (M.Eng.) in Electrical and Computer Engineering from the Aristotle University of Thessaloniki (AUTH). My diploma thesis, “Audio authentication detection through anti-spoofing techniques”, focused on detecting synthetic and manipulated speech. In it, I designed and evaluated three anti-spoofing systems based on the AASIST architecture. Each system used a different front-end representation, ranging from self-supervised models (wav2vec 2.0 XLS-R, Whisper) to classical acoustic features (MFCC). I trained the systems on ASVspoof 2019 LA and evaluated them on five heterogeneous datasets. I built two of these datasets myself to examine how sensitive the models are to noise and to dataset-specific artifacts.

Since 2025 I have been working as a Research Associate at the Centre for Research and Technology Hellas (CERTH), where I take part in European research projects. In the CryptoAction project I developed Python-based tools for blockchain clustering, cryptocurrency analysis and data visualization. I also contribute to the Horizon Europe project O-CEI, which is developing an open Cloud-Edge-IoT platform to support energy flexibility and smart grid pilots across Europe.

My research interests are machine learning, deep learning and artificial intelligence. I focus on audio authentication, multimedia signal processing and biomedical data analysis. I also have experience with large language models and natural language processing. In both my thesis and my work as a Research Associate at CERTH, I have processed large datasets. I have also designed, trained and evaluated AI models across the full pipeline, from data preprocessing to the analysis and visualization of results. My research in audio analysis led to the paper “From benchmarks to the wild: A comparative study of wav2vec 2.0, Whisper and MFCC front-ends in AASIST-based Audio Authentication” (Kostopoulou, Vryzas, Vrysis & Dimoulas), which is currently under review in Speech Communication (Elsevier).

I program in Python, Java, C, C++, MATLAB and SQL, and I use LaTeX.

I hold the Certificate of Proficiency in English (Cambridge and Michigan) and the Goethe-Zertifikat B2 in German. Alongside my studies, I worked as a primary school tutor and as a children’s events animator. These roles strengthened my teamwork, communication and teaching skills.